Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S268446AbUILFDf (ORCPT ); Sun, 12 Sep 2004 01:03:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S268455AbUILFDf (ORCPT ); Sun, 12 Sep 2004 01:03:35 -0400 Received: from fw.osdl.org ([65.172.181.6]:7333 "EHLO mail.osdl.org") by vger.kernel.org with ESMTP id S268446AbUILFCE (ORCPT ); Sun, 12 Sep 2004 01:02:04 -0400 Date: Sat, 11 Sep 2004 22:00:03 -0700 From: Andrew Morton To: Zwane Mwaikambo Cc: torvalds@osdl.org, paulus@samba.org, linux-kernel@vger.kernel.org, anton@samba.org, jun.nakajima@intel.com, ak@suse.de, mingo@elte.hu Subject: Re: [PATCH] Yielding processor resources during lock contention Message-Id: <20040911220003.0e9061ad.akpm@osdl.org> In-Reply-To: References: <16703.60725.153052.169532@cargo.ozlabs.ibm.com> X-Mailer: Sylpheed version 0.9.7 (GTK+ 1.2.10; i386-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 919 Lines: 24 Zwane Mwaikambo wrote: > > The following patch introduces cpu_lock_yield which allows architectures > to possibly yield processor resources during lock contention. err.. Haven't you just invented a really sucky semaphore? > The original > requirement stems from Paul's requirement on PPC64 LPAR systems to yield > the processor to the hypervisor instead of spinning. Maybe Paul needs to use a semaphore. Now, maybe Paul has tied himself into sufficiently tangly locking knots that in some circumstances he needs to spin on the lock and cannot schedule away. But he can still use a semaphore and spin on down_trylock. Confused by all of this. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/